HMD Eye-Region Shadow Detection for Accurate Pupil Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Shadows caused by the configuration and positioning of head-mounted devices (HMDs) interfere with accurate eye-tracking by reducing contrast and brightness differences between the pupil and surrounding eye regions, particularly when shadows appear in the pupil area.
Innovation Solution
A method and device for detecting shadows in the eye region of a user wearing an HMD by determining brightness levels in specific subareas of the eye region, comparing these levels to a predetermined threshold, and generating a signal to indicate the presence of a shadow, thereby improving eye-tracking accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If the HMD configuration and positioning are optimized for display, then the display quality is improved, but shadows appear in the eye region reducing eye-tracking accuracy
Solution Approach 1:
The patent divides the eye region into multiple subareas (pupil area, iris area, sclera area) and independently analyzes brightness levels in each subarea. This segmentation allows the system to detect shadows by comparing brightness differences between specific regions, enabling accurate shadow detection even when overall illumination varies due to HMD configuration optimizations.
Solution Approach 2:
The patent applies local quality analysis by examining brightness characteristics in specific subareas of the eye region rather than the entire eye area. By focusing on local brightness differences between the pupil area and surrounding areas, the system can detect shadows that locally reduce contrast without being affected by global illumination changes from HMD display optimizations.
2Illumination intensity
If illuminators are positioned to provide adequate lighting, then visibility is improved, but shadows are cast in the pupil area reducing contrast
Solution Approach 1:
The patent converts the harmful effect of shadows into a useful detection signal. By analyzing brightness differences between subareas, the system identifies shadow presence as a detectable pattern rather than treating it as mere noise. The shadow-induced brightness reduction in the pupil area becomes a diagnostic feature that triggers shadow detection and correction procedures.
Solution Approach 2:
The patent changes the analysis parameter from absolute brightness levels to relative brightness differences between subareas. This parameter transformation allows the system to detect shadows by comparing the brightness ratio between pupil and surrounding areas, making the detection robust against variations in overall illumination intensity from different illuminator configurations.
3Adaptability or versatility
If cameras are mounted for eye-tracking, then eye movement detection capability is provided, but shadow formation in the eye region increases
Solution Approach 1:
The patent performs preliminary shadow detection by analyzing brightness differences in captured eye region images before proceeding with eye-tracking analysis. This preliminary action identifies shadow conditions that would degrade tracking accuracy, allowing the system to take corrective measures or flag degraded performance before eye movement analysis is compromised.
Solution Approach 2:
The patent implements feedback by using the detected shadow information to adjust or correct the eye-tracking process. When shadows are detected through brightness analysis, the system can compensate for the degraded contrast or indicate reduced tracking reliability, creating a feedback loop that maintains tracking quality awareness based on real-time shadow detection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a method for detecting a shadow (110) in an image (100) of an eye region of a user wearing a Head Mounted Device, HMD. The method comprises obtaining (202), from a camera of the HMD, an image (100) of the eye region of the user wearing a HMD and determining (204) an area of interest (130) in the image (100), the area of interest (130) comprising a plurality of subareas. The method further comprises determining (206) a first brightness level for a first subarea (140) of the plurality of subareas and determining (208) a second brightness level for a second subarea (150, 160) of the plurality of subareas. The method further comprises comparing (210) the first brightness level with the second brightness level, and, based on the comparing, selectively generating (212) a signal indicating a shadow.